6.6.8.5.1 Huge Integer and Subnormal Exceptions
In addition to the IEEE 754-2008/2019 compliant exception support, this macro also offers two additional exceptions and associated masks that some users may find useful.
- Huge Integer: FSR.HUGI
Exception signaled whenever a Float-to-Integer conversion operation (FF2DI and FF2LI) results in an integer value that is larger than the destination register can represent.
- Subnormal Operand:
FSR.SUBO
Exception signaled whenever an operand of an affected instruction is a subnormal value and Subnormals-Are-Zeros (SAZ) mode is disabled (FCR.SAZ = 0). This is the only exception that can be triggered by an operand source condition (all others are related to result conditions).
| Exception | FSR Bit Name | Default Result | ||
|---|---|---|---|---|
| Invalid | INVAL(2) |
Distinguished qNaN or quieted sNaN or Largest integer result (for FF2DI/FF2LI only) | ||
| Divide By Zero | DIV0 | Correctly signed Infinity(3) | ||
| Overflow | OVF | Rounding Mode | Nearest (Even) | Infinity with sign of exact result |
| Zero | Most positive finite number with sign of exact result | |||
| +Infinity |
Positive overflow: +Infinity Negative overflow: Most negative finite number | |||
| -Infinity |
Positive overflow: Most positive finite number Negative overflow: -Infinity | |||
| Underflow | UDF(1) | FCR.FTZ = 0: Rounded
subnormal result | ||
FCR.FTZ = 1; Zero with
sign of exact result | ||||
| Inexact | INX | Rounded (inexact) result | ||
| Huge Integer | HUGI | Largest integer value with sign of input operand | ||
| Subnormal Operator | SUBO | N/A (input operand exception) | ||
|
Note:
| ||||